JP6183209B2 - Video processing apparatus, video processing method, and video processing program - Google Patents

Description

  The present invention relates to a video processing apparatus, a video processing method, and a video processing program.

  With the increasing interest in security, surveillance cameras installed in shopping streets and office buildings are increasing. For example, the security administrator can find a suspicious person by confirming the video of the surveillance camera.

  The video from the surveillance camera is simply stored in the recording device and not used unless an incident or accident occurs. For this reason, for example, as in Patent Documents 1 to 4, techniques related to reduction of the recording amount of video have been proposed.

JP 2007-243699 A International Publication No. 2006/132029 JP 2011-78080 A JP 2004-200909 A

  When an incident or accident occurs, the video of the surveillance camera is used, for example, for specifying a person's movement route. However, the techniques described in Patent Documents 1 to 4 do not effectively reduce the storage amount when specifying the movement path of a person is considered.

  For example, Patent Document 1 describes that the frame rate of a video of a surveillance camera in which a person moving on a specific route is detected is higher than the frame rate of a video of another surveillance camera. However, with this technology, the person being tracked is continuously displayed on the video of multiple surveillance cameras, and even if tracking is easy, the video frame rate is increased, so the video storage capacity is more than necessary. To increase.

  Therefore, the present invention has been made in view of the above-described problems, and an object thereof is to provide a video processing device, a video processing method, and a video processing program that can effectively reduce the recording amount of video.

  The video processing apparatus described in the present specification is obtained by a storage unit that stores map information indicating a target area and a non-target area for imaging, an acquisition unit that acquires videos captured in the target area, and the acquisition unit. A specific processing unit that specifies a movement source area of the subject shown in the captured video based on the map information, and when the area specified by the specific processing unit is the target area, the video of the subject is displayed. A recording processing unit for recording with a lower image quality than when the area specified by the specifying processing unit is the non-target area.

  The video processing method described in the present specification includes a step of acquiring a video shot in a shooting target area, and a moving source area of a subject reflected in the acquired video, and the shooting target area and the non-target area. A step of specifying on the basis of the map information shown, and a step of recording an image of the subject with a lower image quality than when the specified region is the non-target region when the specified region is the target region. Is a method that the computer executes.

  The video processing program described in this specification acquires a video shot in a target area for shooting, and maps the movement source area of the subject reflected in the acquired video to the target area and the non-target area for shooting. If the identified area is the target area, the video is recorded with a lower image quality than the case where the identified area is the non-target area. It is a program to be executed.

  The amount of recorded images can be effectively reduced.

It is a block diagram which shows an example of the monitoring system. It is a map which shows the example of arrangement | positioning of a surveillance camera. It is a block diagram of the video processing apparatus which concerns on an Example. It is a block diagram which shows an example of the function of a video processing apparatus. It is a map which shows the example of the movement path | route of a person. It is a flowchart which shows an example of the registration process of person data. It is a table | surface which shows an example of video data. It is a table | surface which shows an example of the registration content of the person database in an imaging area. It is a table | surface which shows the other example of the registration content of the person database in an imaging area. It is a table | surface which shows an example of a map database. It is a table | surface which shows an example of a person database in a non-photographing area. It is a flowchart (1/2) which shows an example of a video recording process. It is a flowchart (2/2) which shows an example of a video recording process. It is a flowchart which shows an example of the determination process of the resolution according to the similarity between persons. It is a figure which shows an example of the comparison process of the pattern of the clothing according to the resolution. It is a graph which shows the change of the frequency with respect to the brightness | luminance in medium resolution and high resolution. It is a table | surface which shows an example of a resolution setting table. It is a table | surface which shows the other example of a resolution setting table. It is a flowchart which shows an example of the update process of a resolution setting table.

  FIG. 1 is a configuration diagram illustrating an example of a monitoring system. The monitoring system includes a plurality of monitoring cameras (imaging devices) 2, a video processing device 1, and a video storage device 3. The plurality of surveillance cameras 2 are arranged in a plurality of areas, respectively, and take images in the areas.

  FIG. 2 is a map showing an arrangement example of the monitoring camera 2. In this example, each surveillance camera 2 is usually installed in a place (for example, a server room) where only a specific person is permitted to enter, but the present invention is not limited to this. The map information shown in FIG. 2 is held in advance in the video processing device 1 as will be described later.

  In FIG. 2, the insides of the solid line frames A to D indicate the areas where the surveillance cameras 2 are arranged, that is, the areas to be imaged (hereinafter referred to as “imaging areas”). Further, the inside of the dotted frame α indicates an area where the surveillance camera 2 is not arranged, that is, a non-photographing area (hereinafter referred to as “non-photographing area”). Note that the non-photographing area α may be an area where photographing is not performed for reasons such as failure even if the surveillance camera 2 is arranged.

  The overall planar size of the place shown in FIG. 2 is, for example, 6 (m) × 8 (m). The imaging areas A and B exist at the end of the passage L1 extending in the x direction (see the x axis), and the non-imaging area α is an intersection of the passage L1 and the passage L3 extending in the y direction (see the y axis). Exists in the containing area. The shooting area C exists in an area including an intersection of the path L3 and the path L2 extending in the x direction, and the shooting area D exists in an area adjacent to the shooting area C in the path L2.

  In the following description, the position in the place shown in FIG. 2 is indicated according to the x axis and the y axis with the corner of the imaging area A as the origin. Further, the angle θ is indicated with the direction of the x-axis being 0 °.

  Each surveillance camera 2 transmits video data captured in each of the areas A to D to the video processing device 1 via a network NW such as a LAN (Local Area Network). In the present embodiment, a human (person) is cited as an image subject, but the present invention is not limited to this, and another animal such as a dog or a cat may be the subject.

  The video processing device 1 is a computer device, analyzes the video received from each surveillance camera 2, adjusts the video quality according to the movement path of the person who is the subject, and sends the video data to the video storage device 3. Output. As will be described later, the video processing device 1 specifies the area from which the subject is moved based on the map information, and when the specified area is a shooting area, the specified area does not take a picture of the subject. The image is recorded in the video storage device 3 with lower image quality than that in the area. As a result, the amount of video recording is reduced. In the embodiment described below, the resolution is given as an example of the image quality of the video. However, the present invention is not limited to this, and it may be a frame rate.

  The video storage device 3 is, for example, a storage device that stores video data. The video storage device 3 stores the video by writing the video data input from the video processing device 1 to a hard disk drive (HDD). Also, the video stored in the video storage device 3 contributes to the tracking of a suspicious person or the like by being played back, for example, when an incident occurs.

  As described above, the monitoring system of the present example stores the images shot by the plurality of monitoring cameras 2 in the shooting areas A to D with the image quality corresponding to the movement path of the person. As a result, the image recording amount is effectively reduced while maintaining the image quality necessary for tracking the suspicious person.

  Next, the configuration of the video processing device 1 will be described. FIG. 3 is a configuration diagram of the video processing device 1 according to the embodiment.

  The video processing device 1 is a computer device such as a server device. The video processing apparatus 1 includes a CPU (Central Processing Unit) 10, a ROM (Read Only Memory) 11, a RAM (Random Access Memory) 12, an HDD 13, a communication processing unit 14, a portable storage medium drive 15, an input processing unit 16, And an image processing unit 17 and the like.

  The CPU 10 is arithmetic processing means, and executes a video processing method to be described later according to a video processing program. The CPU 10 is connected to the units 11 to 17 via the bus 18. The video processing apparatus 1 is not limited to one that operates by software, and hardware such as an application specific integrated circuit may be used instead of the CPU 10.

  The RAM 12 is used as a working memory for the CPU 10. The ROM 11 and the HDD 13 are used as storage means for storing a video processing program for operating the CPU 10. The communication processing unit 14 is, for example, a network card, and is a communication unit that communicates with an external device such as the monitoring camera 2 or the video storage device 3 via a network such as a LAN.

  The portable storage medium drive 15 is a device that writes information to and reads information from the portable storage medium 150. Examples of the portable storage medium 150 include a USB memory (USB: Universal Serial Bus), a CD-R (Compact Disc Recordable), and a memory card. Note that the video processing program may be stored in the portable storage medium 150.

  The video processing apparatus 1 further includes an input device 160 for performing an information input operation and a display 170 for displaying an image. The input device 160 is input means such as a keyboard and a mouse, and input information is output to the CPU 10 via the input processing unit 16. The display 170 is image display means such as a liquid crystal display, and displayed image data is output from the CPU 10 to the display via the image processing unit 17. In addition, it can replace with the input device 160 and the display 170, and can also use devices, such as a touchscreen provided with these functions.

  The CPU 10 executes a program stored in the ROM 11 or the HDD 13 or a program read from the portable storage medium 150 by the portable storage medium drive 15. This program includes not only the OS (Operating System) but also the video processing program described above. The program may be downloaded via the communication processing unit 14.

  When the CPU 10 executes the video processing program, a plurality of functions are formed. FIG. 4 is a configuration diagram illustrating an example of functions of the video processing device 1. More specifically, FIG. 4 shows an example of functions formed in the CPU 10 and information stored in the HDD 13.

  The CPU 10 includes a video input processing unit (acquisition unit) 100, an analysis processing unit 101, a movement destination area specific low processing part 102, a new registration detection part 103, and a movement source area specific processing part (specific processing part) 104. A recording processing unit 105 and a similarity determination processing unit 106. The recording processing unit 105 includes a resolution determination processing unit 105a and an output processing unit 105b.

  In addition, the HDD (storage unit) 13 includes video data 130, a person database (DB: Data Base) 131 in a shooting area, a person database (DB) 132 in a non-shooting area, a resolution setting table 133, and a map database ( DB) (map information) 134 is stored.

  The outline of the function of each unit will be described. The video input processing unit 100 acquires the video (video data 130) taken by the surveillance camera 2 in the shooting areas A to D. The analysis processing unit 101, the destination area specific low processing unit 102, the new registration detection unit 103, and the movement source area specific processing unit 104 use the map DB 134, the shooting area person DB 131, and the non-shooting area person DB 132, Analyzes the movement path of the person who is the subject.

  The resolution determination processing unit 105 a determines the resolution of the video of the person for each person according to the movement path, and sets the resolution in the resolution setting table 133. The output processing unit 105 b records the video data 130 by converting the video data 130 into a resolution according to the resolution setting table 133 and outputting the converted data to the video storage device 3.

  Next, taking the movement route of three persons as an example, the video processing method according to the above configuration will be described. FIG. 5 is a map showing an example of a person's movement route.

  In FIG. 5, symbols P, Q, and R represent movement paths of the persons X, Y, and Z (identified by solid lines, broken lines, and alternate long and short dash lines), respectively. Reference numerals P1 to P4, Q1 to Q4, and R1 to R4 indicate positions at which the persons X, Y, and Z have entered the shooting areas A to D, respectively, and have left the areas. The entry time and the exit time are appended to the symbols P1 to P4, Q1 to Q4, and R1 to R4.

  The person X enters the shooting area A at time 9:20:00 (see P1), leaves the shooting area A at time 9:20:10 (see P2), and enters the non-shooting area α. . Then, the person X enters the shooting area B at time 9:20:30 (see reference P3), and exits from the shooting area A at time 9:20:40 (see reference P4).

  The person Y enters the shooting area B at time 9:20:15 (see reference Q1), leaves the shooting area B at time 9:20:20 (see reference Q2), and enters the non-shooting area α. . Then, the person Y enters the shooting area C at time 9:20:35 (see reference Q3) and exits from the shooting area A at time 9:20:40 (see reference Q4).

  Thus, the persons X and Y exist in the same non-photographing area α in a common time zone (9:20:20 to 9:20:30). Therefore, considering the tracking of the persons X and Y by video, if the persons X and Y are similar to each other, it is difficult to distinguish each other when they pass through the non-shooting area α and appear again in the shooting areas B and C. It becomes.

  Therefore, the video processing apparatus 1 records the video of the persons X and Y with the lowest resolution that can distinguish the persons X and Y. More specifically, as an example, the video processing device 1 determines the degree of similarity between the clothing patterns Xa and Ya of the persons X and Y by the similarity determination processing unit 106 until the degree of similarity becomes larger than a certain value. Increase the resolution of the video and record the video. The degree of similarity (similarity described later) is an index of similarity, and indicates a higher value as the objects are similar to each other. In addition, the example of the movement path | route of the person X, Y mentioned above is only described with "example 1" in subsequent description.

  On the other hand, the person Z enters the shooting area D at time 9:10:00 (see reference R1), and exits from the shooting area D at time 9:10:10 (see reference R2). Then, the person Z enters the shooting area C at time 9:10:11 (see reference R3), and exits from the shooting area C at time 9:10:17 (see reference R4). Note that the video processing apparatus 1 distinguishes persons by assigning person IDs 001, 002,000 to the persons X, Y, and Z, respectively, in a registration process described later.

  In this manner, the person Z moves from the shooting area D to the shooting area C alone without passing through the non-shooting area α. Accordingly, the tracking of the person Z by the video is easier than the tracking of the persons X and Y.

  Therefore, the video processing apparatus 1 records the video of the person Z at a lower resolution than the video of the persons X and Y. Note that an example of the movement route of the person Z is simply referred to as “example 2” in the following description.

  In order to distinguish between the cases of Example 1 and Example 2 described above, the video processing apparatus 1 identifies the movement path of the persons X, Y, and Z shown in the video from the video and shoots the data of the persons X, Y, and Z Register in the person DB 131 in the area or the person DB 132 in the non-photographing area. Details of the processing will be described below based on the configuration of FIG.

  FIG. 6 is a flowchart illustrating an example of person data registration processing. First, the video input processing unit 100 converts the video data input from each surveillance camera 2 via the communication processing unit 14 into image data in units of frames (step St1). That is, the video input processing unit 100 acquires videos shot in the shooting areas A to D. Next, the video input processing unit 100 writes the converted video data 130 in the HDD 13 (step St2).

  FIG. 7 shows an example of the video data 130. The video data 130 includes a shooting time (see “Time”) and an image (one frame) of one frame at the shooting time.

  The video data 130 of FIG. 7 shows each image of the person Z photographed in the photographing area A in Example 2. According to the video data 130, it can be seen that the person Z appears at the time 9:10: 00 and exits the shooting area A at the time 9:10:10.

  Next, the analysis processing unit 101 analyzes the video data 130 to determine whether or not a person in the video has been detected (step St3). That is, the analysis processing unit 101 determines whether a person is shown in the video. As a person detection means, Japanese Patent Laid-Open No. 2005-15759, or “collaborative object tracking using the relationship between a tracking target and a surrounding area” (Takayoshi Yamashita and other symposiums on image recognition and understanding (MIRU2011) ), Pp. 56-63, 2011).

  If no person is detected (No in step St3), the analysis processing unit 101 performs a process in step St6 described later.

  When a person is detected (Yes in step St3), the analysis processing unit 101 performs the tracking process for the person by analyzing the video data 130 (step St4). At this time, the analysis processing unit 101 recognizes the feature of the person from the image of the video data 130, and detects the position and moving direction of the person X, Y, Z for each time based on the feature. The analysis processing unit 101 updates the photographing area person DB 131 based on the result of the tracking process (step St5).

  FIG. 8 shows an example of registered contents of the person-in-photographer database. FIG. 8 shows the in-photographing area person database DB 131 in which the person Z in Example 2 is registered. FIG. 8 shows the in-photographing area person database DB 131 in two upper and lower stages.

  “Item No.” is a number given in the order of registration. “Person ID” is an identification number for each person. In the case of the person Z, the person ID is “000”.

  “Area ID” is an identification number of a photographing area where a person exists, and in this example, the code (A to D) of each area shown in FIG. In Example 2, since the person Z first appears in the shooting area D, the item No. The area ID of 1 indicates “D”. Since the person Z then appears in the shooting area C, the item No. The area ID of 2 indicates “C”.

  “Appearance time” indicates the time when a person appears in the area. Item No. 1 shows the time (9:10:00) when the person Z appeared in the shooting area D. The appearance time 2 indicates the time (9:10:11) when the person Z appears in the shooting area C.

  “Exit time” indicates the time at which the person left the area. Item No. 1 indicates the time (9:10:10) when the person Z has left the shooting area D. The exit time 2 indicates the time (9:10:17) when the person Z exits the shooting area C. The exit time is updated in step St7 described later.

  “Feature amount” indicates a parameter (index) used for tracking a person. In this example, the “color”, “height”, and “movement speed” of the person are given as the feature quantities. “Color” is represented by numerical values of hue, brightness, and saturation obtained from a person (for example, clothes) in the image. “Height” is calculated from the size of the person in the image. The “movement speed” is calculated, for example, by obtaining the movement distance of the person by comparing the image with the image at the immediately preceding time.

  “Movement path information” indicates the position and movement direction of a person obtained from the image at each time in the video data 130. In the processing of step St4, the analysis processing unit 101 stores the acquired information in each of the columns “1”, “2”,... Each time “movement route information” is acquired from the image for each time (see FIG. 7). , “10” are sequentially recorded.

  The “movement path information” includes “time”, “position”, “movement direction”, and “position on the image”. “Time” is the time of the image and corresponds to the “time” (shooting time) shown in FIG. The “position” is defined based on the x-axis and the y-axis in FIG. 2 and specified from the “position on the image” of the person. Here, “position on the image” is a range of coordinates of an image in which a person is shown. The “movement direction” is defined by the angle θ in FIG. 2 and is specified from the relationship between the face direction of the person or the position of the person shown in the image at the immediately preceding time.

  Further, the analysis processing unit 101 uses the above-described “feature amount” in order to specify “time”, “position”, “movement direction”, and “position on the image”. In this way, the person tracking process is performed. In addition, as a more specific person tracking means, those described in the above Japanese Patent Application Laid-Open No. 2005-15759 or “cooperative object tracking using the relationship between the tracking target and the surrounding area” can be cited. It is done.

  “Resolution designated value” (K) indicates the resolution determined by the resolution determination processing unit 105a, as will be described later. In this embodiment, the resolution is handled as “low resolution” (K = 1), “medium resolution” (K = 2), and “high resolution” (K = 3). The “resolution designation value” is set to “low resolution” (K = 1) as an initial value. The resolution determination method will be described later.

  FIG. 9 shows another example of the registered contents of the person-in-area database DB 131. FIG. 9 shows an in-photographing area person database DB 131 in which persons X and Y in Example 1 are registered. FIG. 9 shows the in-photographing area person database DB 131 in two upper and lower stages.

  Item No. 1 is registration information of the person X in the photographing area A. Reference numeral 2 denotes registration information of the person Y in the shooting area B. Item No. 3 is registration information of the person X in the photographing area B. 4 is registration information of the person Y in the shooting area C.

  Referring to FIG. 6 again, the analysis processing unit 101 determines whether there is a person who has left the shooting area (step St6). At this time, the analysis processing unit 101 determines that there is a person who has left when it detects a person that exists in the image at the previous time and does not exist in the image (latest image).

  If no person has left (No in step St6), the process in step St11 described later is performed. If there is a person who has left, Yes in step St6), the analysis processing unit 101 writes the leaving time of the person-in-area DB 131 (step St7).

  Next, the movement destination area identification processing unit 102 identifies the movement destination area of the person who has left the office based on the map DB 134 (step St8). The map DB 134 uses the map shown in FIG. 2 as numerical information. The destination area refers to the area immediately after the time series.

  FIG. 10 shows an example of the map DB 134. The map DB 134 includes “area ID”, “monitoring presence / absence” information, “position” information, and “connection area” information, and indicates the imaging areas A to D and the non-imaging area α in FIG. “Area ID” is identification information of each area and corresponds to the code (A to D, α) of each area shown in FIG.

  The “monitoring presence / absence” information indicates whether or not the surveillance camera 2 is shooting the area (see “Yes”) or not (see “No”). That is, the “monitoring presence / absence” information indicates whether the area is a photographing area (“present”) or a non-photographing area (“none”). For this reason, the “monitoring presence / absence” information of the areas A to D indicates “present”, and the “monitoring presence / absence” information of the area α indicates “none”. Note that the area where the “monitoring presence / absence” information is “none” (in this example, the area α) is not only the area where the monitoring camera 2 is not provided, but even if the monitoring camera 2 is present, photographing is performed for reasons such as failure. This includes areas that are not.

  The “position” information indicates the range occupied by the area by the positions of two points facing each other on the diagonal of the rectangle of the area. The positions of the two points are defined by the x axis and the y axis in FIG.

  The “connection area” information indicates an area ID of an adjacent area connected to the area through the passages L1 and L2. The area ID of the adjacent area is registered in each column “1”, “2”, “3”,.

  Since the imaging areas A and B are adjacent to the non-imaging area α by the passage L1, the “connection area” information indicates “α”. Since the imaging area C is adjacent to the non-imaging area α by the passage L3 and adjacent to the non-imaging area D by the passage L2, the “connection area” information indicates “α” and “D”.

  Since the imaging area D is adjacent to the imaging area C by the passage L2, the “connection area” information indicates “C”. Since the non-photographing area α is adjacent to the photographing areas A and B by the passage L1 and adjacent to the photographing area C by the passage L3, the “connection area” information indicates “A”, “B”, and “C”. .

  Thus, since the map DB 134 shows the connection relationship between the areas A to D and α, the destination area specifying processing unit 102 can recognize the map shown in FIG. 2 by referring to the map DB 134.

  At this time, the movement destination area identification processing unit 102 acquires the final movement route information of the person who has left from the person-in-area DB 131. The movement destination area identification processing unit 102 identifies the movement destination area based on the map DB 134 from the acquired final movement route information.

  In the case of Example 2, the movement destination area identification processing unit 102 obtains the final movement path information (movement path information of item No. 1) of the person Z who has left the shooting area D from the shooting area person DB 131 shown in FIG. “10”). The final movement route information of the person Z indicates that the “position” is (300, 400) and the “movement direction” is 0 °. Accordingly, the movement destination area identification processing unit 102 identifies the movement destination of the person Z as the shooting area C by referring to the map DB 134.

  In the case of Example 1, the movement destination area identification processing unit 102 stores the final movement path information (movement path information of item No. 1) of the person X who has left the shooting area A from the shooting area person DB 131 shown in FIG. “10”). The final movement route information of the person X indicates that the “position” is (200, 100) and the “movement direction” is 0 °.

  Accordingly, the movement destination area identification processing unit 102 identifies the movement destination of the person X as the non-photographing area α by referring to the map DB 134. Similarly, the movement destination area identification processing unit 102 acquires the final movement path information (movement path information “10” of item No. 2) of the person Y who has left the shooting area B, and refers to the map DB 134. The destination of the person Y is identified as the non-photographing area α.

  Referring to FIG. 6 again, after specifying the destination, the destination area specifying unit 102 determines whether or not the specified destination is the non-imaging area α (step St9). When the destination is the shooting areas A to D (No in step St9), the destination area specifying unit 102 performs the process in step St11 described later.

  If the destination is the non-photographing area α (Yes in step St9), the destination area specifying processing unit 102 registers the person ID of the person who has left in the non-photographing area person DB 132 (step St10). The non-photographing area person DB 132 indicates person IDs of persons existing in the non-photographing area α.

  FIG. 11 shows an example of the person DB 132 in the non-photographing area. The non-photographing area person DB 132 includes “area ID”, “person ID”, and “exit number”. “Section ID” indicates a section ID (α in this example) of a non-photographing section.

  “Person ID” corresponds to the above-described person ID and indicates an identification number of a person existing in the area. In the case of Example 2, since the persons X and Y move from the shooting areas A and B to the non-shooting area α, the person IDs “001” and “002” of the persons X and Y are registered as the “person ID”. The

  The “number of people leaving” is added every time a person leaves the area in the video recording process described later. As will be described later, when the “number of people leaving” reaches the number of “person ID”, the similarity determination processing unit 106 determines that there is no person in the non-photographing area, and determines the degree of similarity between persons. judge. That is, in the case of Example 2, when the “number of people leaving” is 2, the similarity determination processing unit 106 determines the degree of similarity between the persons X and Y.

  After the person ID is registered in the non-photographing area person DB 132, if the process is not terminated (No in step St11), the process in step St1 is executed again. If not (Yes in step St11), the process is terminated. In this way, the person data registration process is performed.

  Next, video recording processing will be described. In this process, the person-in-photographing area DB 131 is used to specify the video image of the person after determining the resolution for each person. In addition, the non-photographing area person DB 132 is used to determine whether or not all persons have left the non-photographing area α.

  12A and 12B are flowcharts illustrating an example of video recording processing. FIG. 12A and FIG. 12B show a flowchart of this process by being connected at circled positions “A” to “C” described in FIG. 12A and FIG. 12B.

  First, the new registration determination unit 103 reads out the in-photographing area person DB 131 (step St21). Next, the new registration determination unit 103 determines the presence / absence of a person newly registered in the person-in-shooting area DB 131 (step St22). That is, the new registration determination unit 103 determines whether or not information on a person who has newly appeared in the shooting area is registered in the shooting area person DB 131.

  When there is no newly registered person (No in step St22), that is, when there is no person newly appearing in the shooting areas A to D, the resolution determination processing unit 105a is based on the resolution designation value in the shooting area person DB 131. Then, the resolution setting table 133 is updated (step St38). The resolution determination processing unit 105a reflects the resolution determined for each person in the resolution designation value in the process of step St29 described later. As will be described later, the resolution setting table 133 indicates the resolution at each shooting time (see FIGS. 16 and 17). Then, the process of step St33 mentioned later is performed.

  When there is a newly registered person (Yes in step St22), the movement source area identification processing unit 104 identifies the movement source area of the person based on the map DB 134 (step St23). That is, the movement source area identification processing unit 104 identifies the movement source area of the person shown in the video acquired by the video input processing unit 100 based on the map DB 134. The source area refers to the area immediately before in time series.

  At this time, the movement source area identification processing unit 104 acquires the first movement route information of the person who has left from the person-in-area DB 131. The movement source area identification processing unit 104 identifies a movement source area based on the acquired map DB 134 from the acquired first movement route information.

  In the case of Example 2, the movement source area identification processing unit 104 stores the first movement path information (movement path information of item No. 2) of the person Z that appears in the shooting area C from the shooting area person DB 131 shown in FIG. “1”) is acquired. The final movement route information of the person Z indicates that the “position” is (300, 400) and the “movement direction” is 0 °. Accordingly, the movement destination area identification processing unit 102 identifies the movement source of the person Z as the imaging area D by referring to the map DB 134.

  In the case of Example 1, the movement source area identification processing unit 104 reads the first movement path information (movement path information of item No. 3) of the person X that appears in the shooting area B from the in-capture area person DB 131 shown in FIG. “1”) is acquired. The first movement route information of the person X indicates that the “position” is (600, 100) and the “movement direction” is 0 °.

  Therefore, the movement destination area identification processing unit 102 identifies the movement source of the person X as the non-photographing area α by referring to the map DB 134. Similarly, the movement source area identification processing unit 104 acquires the first movement path information (movement path information “1” of item No. 4) of the person Y who appears in the shooting area C, and refers to the map DB 134. The movement source of the person Y is specified as the non-photographing area α.

  When the specified area is the non-photographing area α (Yes in step St24), the movement source area specifying processing unit 104 updates the number of people leaving in the non-photographing area person DB 132 (step St25). That is, the movement source area identification processing unit 104 adds 1 to the number of people leaving the movement source non-photographing area α.

  Next, the resolution determination processing unit 105a determines whether or not the non-photographing area α specified by the movement source area specifying processing unit 104 is unattended by referring to the number of people leaving the non-photographing area person DB 132. (Step St26). That is, the resolution determination processing unit 105a determines whether or not the number of person IDs in the non-photographing area person DB 132 matches the number of people leaving.

  When the non-photographing area α is not unattended (No in step St26), the resolution determination processing unit 105a performs the process in step St38 described above.

  In addition, when the non-photographing area α is unattended (Yes in step St26), the resolution determination processing unit 105a counts the number of person IDs in the non-photographing area person DB 132 so that the number of persons who are in the non-photographing area α can be determined. It is determined whether or not there are a plurality (step St27). When the number of persons in the non-photographing area α is singular (No in Step St27), the resolution determination processing unit 105a determines the resolution K of the person in the non-photographing area α to “2” (medium resolution) (Step S27). St39). Then, the process of step St29 mentioned later is performed.

  As such a case, in Example 1, the person Y does not exist and only the person X appears in the shooting area B through the non-shooting area α. In this case, since the person X alone passes through the non-photographing area α, when the person X in the video is tracked, it is not mistaken for another person similar to the person X. However, since the person X enters the non-photographing area α after passing through the photographing area A, the person X temporarily disappears from the video. For this reason, the resolution determination processing unit 105a makes it easy to track the person X by raising the resolution one step from the initial minimum resolution (K = 1) to the medium resolution (K = 2).

  In addition, when there are a plurality of persons in the non-photographing area α (Yes in step St27), the resolution determination processing unit 105a determines each person according to the degree of similarity between the persons in the non-photographing area α, that is, the similarity. Is determined (step St28). As such a case, as in Example 1, a plurality of persons X and Y exist in the same non-shooting area α in a common time zone (9:20:20 to 9:20:30). Is mentioned. Details of the resolution determination process in this case will be described below.

  FIG. 13 is a flowchart illustrating an example of resolution determination processing according to the similarity between persons. First, the resolution determination processing unit 105a temporarily determines the resolution K of the video of the person in the non-photographing area α to “2” (medium resolution) (step St41). Next, based on the instruction from the resolution determination processing unit 105a, the similarity determination processing unit 106 extracts, from the video data 130, an image capable of comparing persons from the video in which each person is shown (step St42).

  More specifically, the similarity determination processing unit 106 acquires the person ID from the resolution determination processing unit 105a, and based on the acquired person ID, the area ID corresponding to the person ID and the appearance from the person-in-area DB 131. Search for time (or exit time). Then, based on the search result, the similarity determination processing unit 106 extracts an image from which the persons can be compared from the video data 130. As such an image, for example, a combination of images having the same size and orientation (movement direction in FIG. 9) of each person can be cited.

  Next, the similarity determination processing unit 106 converts the extracted image into an image with the determined resolution K (step St43). When the resolution is reduced (K = 2), the similarity determination processing unit 106 divides the image into predetermined sizes, and averages the pixel values in the unit image obtained by the division. For example, when the resolution is 1/9, the entire image is divided into 3 × 3 pixels, and the pixel values in the unit image are averaged.

  Next, the similarity determination processing unit 106 calculates a similarity cor between persons from the extracted image (step St44). At this time, the similarity determination processing unit 106 extracts the feature of each person from the image, and calculates the similarity by comparing the feature amounts. As the feature amount, for example, color information indicating the color of a person, arrangement information indicating the arrangement of colors, pattern information indicating a pattern of clothes, and the like are used. The similarity determination processing unit 106 calculates a similarity cor by comparing feature amounts. The calculation method will be described later.

  Next, when the similarity degree cor is equal to or less than the threshold value (constant value) TH (Yes in step St45), the similarity determination processing unit 106 determines the determined resolution K (step St46). The determined resolution K is reflected in the resolution designation value of the person-in-photographing area DB 131 or the resolution setting table 133.

  Further, when the similarity cor is larger than the threshold (constant value) TH (No in Step St45), the similarity determination processing unit 106 increases the determined resolution K by one level (that is, K = K + 1) (Step St46). In the present embodiment, the resolution K has only three stages “1” to “3”, and the initial value is “2” (see step St41). Therefore, with this process, the resolution K is “3” (high resolution). It becomes. However, the resolution K is not limited to this, and may be, for example, five levels (“1” to “5”). In this case, the resolution K is increased to “5” at the maximum by this processing. .

  Thereafter, the similarity determination processing unit 106 performs the process of step St43 again, and further performs a calculation process and a determination process (steps St44 and St45) of the similarity degree cor. That is, the similarity determination processing unit 106 increases the resolution of an image in which a person is shown until the similarity cor becomes equal to or less than the threshold value TH.

  FIG. 14 shows an example of a clothing pattern comparison process according to the resolution. FIG. 14 compares the patterns of the clothes Xa of the person X and the clothes Ya of the person Y in Example 1 when the resolution K is “2” (medium resolution) and “3” (high resolution). The state of performing is shown.

  As understood from FIG. 14, when the resolution K is “2” (medium resolution), the pattern of the clothes Xa of the person X and the clothes Ya of the person Y have a high degree of similarity cor and are difficult to distinguish from each other. is there. For this reason, the similarity determination processing unit 106 determines that the similarity cor is greater than the threshold value TH in the determination process of step St45. Then, the similarity determination processing unit 106 changes the resolution K to “3” (step St48), and again performs the determination process of the similarity cor (step St45).

  When the resolution K is “3” (high resolution), the patterns of the clothes Xa of the person X and the clothes Ya of the person Y have a low similarity cor and can be distinguished from each other. That is, the clothing Ya of the person Y is a mesh pattern, whereas the clothing Xa of the person X is plain, so that the patterns are different from each other to be distinguishable.

  For this reason, the similarity determination processing unit 106 determines that the similarity cor is equal to or less than the threshold value TH in the determination process of step St45. Then, the similarity determination processing unit 106 determines the determined resolution K (= “3”) (step St46) and ends the process.

  FIG. 15 is a graph showing changes in frequency with respect to luminance at medium resolution and high resolution. FIG. 15A shows frequency change characteristics with respect to the brightness of each pattern of the clothes Xa of the person X and the clothes Ya of the person Y when the resolution K is “2” (medium resolution). FIG. 15B shows frequency change characteristics with respect to the brightness of each pattern of the clothes Xa of the person X and the clothes Ya of the person Y when the resolution K is “3” (high resolution).

  Each change characteristic is similar at medium resolution but distinctly different at high resolution. The similarity determination processing unit 106 can calculate the similarity cor based on the following equation (1) from the change characteristics shown in FIG. 15, that is, the luminance histogram data.

  Here, the functions a (x) and b (x) represent histogram data relating to the persons X and Y, respectively. The variable n is a maximum value within a numerical range that the luminance can take.

  The similarity cor calculated based on the equation (1) can take a value of 0 to 1. Here, the degree of similarity cor = 1 means that the patterns of the clothes Xa and Ya of the persons X and Y are most similar.

  When the threshold value TH = 0.5, if the similarity cor calculated at medium resolution is 0.8 and the similarity cor calculated at high resolution is 0.2, 0.2 ≦ 0.5. The resolution K is determined to be “3” (high resolution). In this way, resolution determination processing is performed.

  As described above, when there is another person who has moved from the non-target area α to the shooting area where the video of the person was shot or to another shooting area, the similarity determination processing unit 106 calculates the similarity cor between the persons, The determination is made based on the video data 130. In addition, when the similarity cor is equal to or lower than the threshold value TH, the recording processing unit 105 has a lower resolution (that is, medium resolution (K = 2)) than the case where the similarity degree cor is larger than the threshold value TH. Record with.

  That is, the similarity determination processing unit 106 increases the resolution K to “3” when the similarity cor calculated at the medium resolution (K = 2) is equal to or less than the threshold value TH (Yes in step St45) (step St48). Then, the resolution K is fixed (step St46). For this reason, the image of the person is recorded in the video storage device 3 at a medium resolution, and the recording amount is reduced as compared with the case of recording at a high resolution. This is because, when a person is tracked by video, tracking is not difficult even for a medium resolution video if there is no other person similar to the person.

  Further, the recording processing unit 105 records an image showing a person with a lower resolution as the similarity cor is lower. That is, the similarity determination processing unit 106 executes the processing for increasing the resolution (K = K + 1, step St48) as the similarity cor is lower. For this reason, the amount of recorded images is effectively reduced.

  In other words, when the similarity cor is larger than the threshold value TH, the recording processing unit 105 records the video by increasing the resolution of the video in which the person is shown so that the similarity cor is equal to or lower than the threshold value TH. That is, the similarity determination processing unit 106 repeats the process of Step St48 (K = K + 1) to increase the resolution as the similarity cor calculated first is higher. As a result, the video is recorded after being adjusted to the optimum resolution for tracking the person.

  Referring to FIGS. 12A and 12B again, the resolution determination processing unit 105a updates the resolution designation value of the person in the photographing area person DB 131 (step St29). In the case of Example 1, the resolution determination processing unit 105a sets the resolution designation value of the persons X and Y to “3” (high resolution) based on the result of the above-described processing (see FIG. 9).

  Next, the resolution determination processing unit 105a updates the resolution setting table 133 (step St30). That is, the resolution determination processing unit 105 a reflects the resolution designation value in the resolution setting table 133.

  FIG. 16 shows an example of the resolution setting table 133. FIG. 16 shows a resolution setting table 133 related to the images of the persons X and Y in Example 1. The resolution setting table 133 includes “time” information corresponding to the shooting time, and “area ID” indicating the shooting area of the video in which the person is shown. The resolution setting table 133 shows the set values of resolution for each time and area. In FIG. 16, the resolution setting value is expressed as “high” when the resolution is high (K = 3), and is expressed as “low” when the resolution is low (K = 1).

  The resolution determination processing unit 105a searches the person ID X, Y area ID, appearance time, and exit time from the person-in-capture person DB 131, and identifies the video of the persons X, Y based on the search result. Then, the resolution determination processing unit 105a updates the resolution setting value according to the shooting time and shooting area of the specified video.

  In the case of Example 2, the resolution determination processing unit 105a sets the video resolution of the persons X and Y to a high resolution. That is, the image of the person X at time 9:20:00 to 9:20:10 in the shooting area A (movement source) and the time 9:20:30 to 9:20:40 in the shooting area B (movement destination). The video of the person X is set to high resolution. Furthermore, the image of the person Y at the time 9:20:15 to 9:20:20 in the shooting area B (movement source) and the time 9:20:35 to 9:20:40 in the shooting area C (movement destination). The video of the person Y is also set to high resolution. For other videos, since no person is shown, the initial value is set to “low”.

  On the other hand, in the case of Example 2, the resolution determination processing unit 105a determines the resolution K to be “1” (low resolution) through the processing different from that in Example 1 in Step St24 described above, and sets the resolution setting table 133 in the resolution setting table 133. To do.

  The resolution determination processing unit 105a determines the resolution K of the person to be “1” when the areas specified by the movement source area specifying processing unit 104 are the shooting areas A to D (No in Step St24) (Step St40). . In Example 2, the movement source of the person Z that appears in the shooting area C is specified as the shooting area C as described above. Therefore, the resolution K of the video of the person Z is determined to be “1” (low resolution).

  Next, the resolution determination processing unit 105a updates the resolution designation value of the person in the photographing area person DB 131 (step St29). In the case of Example 2, the resolution determination processing unit 105a sets the resolution designation value of the person Z to “1” (low resolution) based on the result of the above-described processing (see FIG. 8). Next, the resolution determination processing unit 105a updates the resolution setting table 133 (step St30).

  FIG. 17 shows an example of the resolution setting table 133. FIG. 17 shows a resolution setting table 133 related to the video of the person Z in the case of Example 2.

  In the case of Example 1, the resolution determination processing unit 105a sets the video resolution of the person Z to a low resolution. That is, the image of the person Z at the time 9: 10: 0 to 9:10:10 in the shooting area D (movement source) and the time 9:10:11 to 9:10:17 in the shooting area C (movement destination). The video of the person Z is set to a low resolution. For other videos, since no person is shown, the initial value is set to “low”.

  Therefore, when the areas specified by the movement source area specifying processor 104 are the shooting areas A to D, the recording processing unit 105 displays an image showing a person, and the specified area is a non-shooting area α. Record at a lower resolution. That is, as described above, when the specified area is the non-photographing area α (Yes in step St24), the resolution K is determined to be “2” (medium resolution) or “3” (high resolution). . On the other hand, when the specified area is the imaging areas A to D (No in step St24), the resolution K is determined to be “1” (low resolution).

  This is because when the person Z moves from the shooting area D to the shooting area C alone without passing through the non-shooting area α, the tracking of the person Z by the image (example 2) passes through the non-shooting area α. This is because it is easier than tracking similar persons X and Y (example 1). That is, when the person Z moves from the shooting area D to another adjacent shooting area C as in the case of Example 2, the image of the person Z is displayed continuously in the images of the plurality of surveillance cameras 2. Even if the resolution (image quality) is reduced, tracking is sufficiently possible. Therefore, according to the video processing apparatus 1 according to the embodiment, the video recording amount is effectively reduced according to the movement path of the person, unlike the contents described in Patent Document 1 and the like.

  In addition, when the areas specified by the movement source area specification processing unit 104 are the shooting areas A to D (in the case of Example 2), the recording processing unit 105 moves an image in which a person is reflected in the movement source shooting area. Record at the same resolution as the video of the person in the previous shooting area. As a result, the recording amount of the video is reduced as compared with the case where only the resolution of the video in one shooting area of the movement source or the movement destination is lowered.

  Thus, since the resolution setting table 133 is set according to the movement path of the person, the set resolution may be changed later. As such a case, in Example 1, after the person Z moves from the shooting area C to the non-shooting area α, another person similar to the person Z enters the non-shooting area α.

  In this case, as described above, the resolution of the person Z is increased according to the similarity cor between persons (see FIG. 13). For this reason, the resolution (“low” (K = 1)) of the person Z that has already been set may be changed to a high resolution (K = 1) or a medium resolution (K = 2). Therefore, the resolution setting table 133 is updated only when the resolution designation value is larger than the existing setting value.

  FIG. 18 is a flowchart illustrating an example of the update process of the resolution setting table 133. First, the resolution determination processing unit 105a reads an existing setting value from the resolution setting table 133 (step St51). Next, the resolution determination processing unit 105a compares the resolution designation value with the existing setting value (Step St52).

  If the resolution designation value is larger than the existing setting value (Yes in step St52), the resolution determination processing unit 105a writes the resolution designation value in the resolution setting table 133 (step St53), and ends the process. In addition, when the resolution designation value is equal to or less than the existing setting value (No in step St52), the resolution determination processing unit 105a ends the process without writing the resolution designation value. In this way, the update process of the resolution setting table 133 is performed.

  After the resolution setting table 133 is updated, the resolution determination processing unit 105a deletes the registration related to the movement source area of the person from the shooting area person DB 131 (step St31). In the case of Example 1, the resolution determination processing unit 105a obtains the item No. that is registered information of the shooting area A from which the person X is moved from the person DB 131 in the shooting area of FIG. 1 and item No. 1 which is registration information of the shooting area B from which the person Y is moved. 2 information is deleted.

  Further, in the case of Example 2, the resolution determination processing unit 105a acquires the item No. that is registration information of the shooting area D from which the person Z is moved from the in-shooting area person DB 131 of FIG. 1 information is deleted. As a result, the in-photographing area person DB 131 is updated at any time as the person moves.

  Next, the resolution determination processing unit 105a deletes the person ID from the non-photographing area person DB 132 (step St32). In the case of Example 1, the person IDs 001.002 of the persons X and Y are deleted from the person DB 132 in the non-photographing area. Thereby, the person DB 132 in the non-photographing area is updated at any time as the person moves.

  Next, the output processing unit 105b acquires the earliest appearance time from the non-photographing area person DB 132 (step St33). In the case of Example 1, the output processing unit 105b obtains the item number “No. 3 appearance time 09:20:30 is acquired. Further, in the case of Example 2, the output processing unit 105b obtains the item number “No. 2 appearance time 09:10:11 is acquired.

  Next, the output processing unit 105b determines whether or not there is unrecorded video data before the acquired appearance time in the video data 130 (step St34). When there is no unrecorded video data (No in step St34), the output processing unit 105b performs a process in step St36 described later.

  If there is unrecorded video data (Yes in step St34), the output processing unit 105b records the video data with a resolution according to the resolution setting table 133 (step St35). That is, the output processing unit 105 b converts the video data into a resolution according to the setting value of the resolution setting table 133 and outputs the converted data to the video storage device 3.

  As a result, the images of the shooting areas A to D that were immediately before the person moved to the current area are recorded in the video storage device 3 with the determined resolution. That is, in the case of Example 1, when the persons X and Y move to the shooting areas B and C, the images of the persons X and Y shot in the shooting areas A and B are recorded. In the case of Example 2, when the person Z moves to the shooting area C, the video of the person Z shot in the shooting area D is recorded.

  Therefore, the video data 130 is recorded at any time as the person moves. Note that the video recording process is not limited to this, and may be performed every time a predetermined time elapses using a timer, for example.

  Next, the output processing unit 105b records, with low resolution (K = 1), video data in which no person is shown among the video data 130 up to the current time (Step St36). At this time, the output processing unit 105b searches for the time and area where no person is shown based on the person DB 131 in the shooting area, acquires the video data 130 according to the search result, converts it to a low resolution, and converts the video Output to the storage device 3.

  Thereafter, when the process is continued (No in Step St37), the process in Step St21 is performed again. Otherwise (Yes in Step St37), the process is terminated. In this way, video recording processing is performed.

  As described above, the video processing apparatus 1 according to the embodiment includes the storage unit (HDD) 13, the acquisition unit (video input processing unit) 100, the specific processing unit (movement source area specific processing unit) 104, And a recording processing unit 105. The storage unit 13 stores map information (map DB) 134 indicating a target area (photographing area) and a non-target area (non-photographing area).

  The acquisition unit 100 acquires a video shot in the target area. The identification processing unit 104 identifies the movement source area of the subject (person) shown in the video acquired by the acquisition unit 100 based on the map information 134. When the area specified by the specific processing unit 104 is the target area, the recording processing unit 105 displays an image showing the subject with a lower image quality (resolution) than when the area specified by the specific processing unit 104 is a non-target area. ) To record.

  Accordingly, when the subject moves from the shooting target area to another target area, the recording processing unit 105 records the video image of the subject with lower image quality than when the subject moves to the non-target shooting area. Therefore, the video processing apparatus 1 according to the embodiment can effectively reduce the video recording amount.

In the video processing method according to the embodiment, the computer (CPU) 10 executes the following steps.
Step (1): An image shot in a shooting target area (shooting area) is acquired.
Step (2): The movement source area of the subject (person) shown in the acquired video is specified based on the map information (map DB) 134 indicating the shooting target area and the non-target area (non-shooting area).
Step (3): When the specified area is the target area, an image showing the subject is recorded with lower image quality than when the specified area is the non-target area.

  Since the video processing method according to the embodiment has the same configuration as that of the video processing device 1 described above, the same effects as those described above are achieved.

Further, the video processing program according to the embodiment causes the computer (CPU) 10 to execute the following processing.
Process (1): An image shot in a shooting target area (shooting area) is acquired.
Process (2): The movement source area of the subject (person) shown in the acquired video is specified based on the map information (map DB) 134 indicating the shooting target area and the non-target area (non-shooting area).
Process (3): When the specified area is the target area, an image showing the subject is recorded with lower image quality than when the specified area is the non-target area.

  Since the video processing program according to the embodiment has the same configuration as that of the video processing device 1 described above, the same effects as those described above can be obtained.

  The above processing functions can be realized by a computer. In that case, a program describing the processing contents of the functions that the processing apparatus should have is provided. By executing the program on a computer, the above processing functions are realized on the computer. The program describing the processing contents can be recorded on a computer-readable recording medium (except for a carrier wave).

  When the program is distributed, for example, it is sold in the form of a portable recording medium such as a DVD (Digital Versatile Disc) or a CD-ROM (Compact Disc Read Only Memory) on which the program is recorded. It is also possible to store the program in a storage device of a server computer and transfer the program from the server computer to another computer via a network.

  The computer that executes the program stores, for example, the program recorded on the portable recording medium or the program transferred from the server computer in its own storage device. Then, the computer reads the program from its own storage device and executes processing according to the program. The computer can also read the program directly from the portable recording medium and execute processing according to the program. Further, each time the program is transferred from the server computer, the computer can sequentially execute processing according to the received program.

  The above-described embodiment is an example of a preferred embodiment of the present invention. However, the present invention is not limited to this, and various modifications can be made without departing from the scope of the present invention.

In addition, the following additional notes are disclosed regarding the above description.
(Supplementary Note 1) A storage unit that stores map information indicating a target area and a non-target area for shooting;
An acquisition unit for acquiring a video imaged in the target area;
A specific processing unit that identifies a movement source area of the subject shown in the video acquired by the acquisition unit based on the map information;
When the area specified by the specific processing unit is the target area, a recording process for recording the video image of the subject with lower image quality than when the area specified by the specific processing unit is the non-target area And a video processing apparatus.
(Additional remark 2) When the area specified by the specific processing unit is the non-target area, the other area that has moved from the non-target area to the target area where the video of the subject was captured or to the other target area A determination processing unit that determines a degree of similarity between the subject and the other subject based on the video acquired by the acquisition unit when the subject exists;
The recording processing unit, when the degree of similarity is equal to or less than a certain value, records the video image of the subject with lower image quality than when the degree of similarity is larger than the certain value. The video processing apparatus according to 1.
(Supplementary note 3) The video processing apparatus according to supplementary note 2, wherein the recording processing unit records the video in which the subject is reflected with lower image quality as the degree of similarity is lower.
(Supplementary Note 4) When the degree of similarity is larger than the certain value, the recording processing unit increases the image quality of the video in which the subject is captured so that the degree of similarity is equal to or less than the certain value. The video processing apparatus according to appendix 2, wherein the video in which the image is reflected is recorded.
(Supplementary Note 5) When the area specified by the specific processing unit is the target area, the recording processing unit displays an image of the subject in the movement source area and the object in the movement destination area. The video processing apparatus according to any one of appendices 1 to 4, wherein the video processing is recorded with the same image quality as the recorded video.
(Appendix 6) A step of acquiring a video shot in a target area for shooting;
Identifying a movement source area of the subject shown in the acquired video based on map information indicating the target area and the non-target area of shooting;
When the specified area is the target area, the computer executes a step of recording a video image of the subject with a lower image quality than when the specified area is the non-target area. Video processing method.
(Supplementary Note 7) When the specified area is the non-target area, when there is another subject that has moved from the non-target area to the target area where the video of the subject is captured or to the other target area The computer further executes a step of determining the degree of similarity between the subject and the other subject based on the acquired video,
In the step of recording the video showing the subject, if the degree of similarity is below a certain value, the video showing the subject is recorded with a lower image quality than when the degree of similarity is larger than the certain value. The video processing method according to appendix 6, wherein:
(Supplementary note 8) The video processing method according to supplementary note 7, wherein in the step of recording the video image of the subject, the video image of the subject is recorded with lower image quality as the degree of similarity is lower.
(Supplementary Note 9) In the step of recording the video image of the subject, the image quality of the video image of the subject so that the similarity level is less than or equal to the predetermined value when the similarity level is greater than the predetermined value. The video processing method according to claim 7, further comprising: recording a video showing the subject.
(Supplementary Note 10) In the step of recording the video image of the subject, when the specified area is the target area, the video image of the subject in the movement source area is displayed in the movement destination area. 10. The video processing method according to any one of appendices 6 to 9, wherein the video is recorded with the same image quality as the projected video.
(Appendix 11) Obtaining images shot in the shooting area,
Identify the area from which the subject appears in the acquired video based on the map information indicating the target area and non-target area of shooting,
When the specified area is the target area, the computer causes the computer to execute a process of recording a video image of the subject with lower image quality than when the specified area is the non-target area. Video processing program.
(Supplementary Note 12) When the specified area is the non-target area, when there is another subject that has moved from the non-target area to the target area where the video of the subject is captured or to the other target area , Further causing the computer to execute a process of determining the degree of similarity between the subject and the other subject based on the acquired video,
In the process of recording the video showing the subject, when the degree of similarity is less than or equal to a certain value, the video showing the subject is recorded with lower image quality than when the degree of similarity is larger than the certain value. The video processing program according to supplementary note 11, characterized in that:
(Supplementary note 13) The video processing program according to supplementary note 12, wherein in the process of recording the video showing the subject, the video showing the subject is recorded with lower image quality as the degree of similarity is lower.
(Supplementary Note 14) In the process of recording the video image of the subject, the image quality of the video image of the subject so that the similarity level is less than or equal to the predetermined value when the similarity level is greater than the predetermined value. The video processing program according to appendix 12, wherein the video showing the subject is recorded.
(Supplementary Note 15) In the process of recording the video image of the subject, when the specified area is the target area, the video image of the subject in the movement source area is displayed in the movement destination area. 15. The video processing program according to any one of appendices 11 to 14, wherein the video processing program is recorded with the same image quality as the projected video.

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 2 Surveillance camera 10 CPU
13 HDD (storage unit)
100 Video input processing unit (acquisition unit)
104 Source area identification processing unit (specific processing unit)
105 Recording processing unit 106 Determination processing unit (similarity determination processing unit)
130 Video data 134 Map database (map information)

Claims (7)

A storage unit that stores map information indicating a target area and a non-target area of shooting;
An acquisition unit for acquiring a video imaged in the target area;
A specific processing unit that identifies a movement source area of the subject shown in the video acquired by the acquisition unit based on the map information;
When the area specified by the specific processing unit is the target area, a recording process for recording the video image of the subject with lower image quality than when the area specified by the specific processing unit is the non-target area And a video processing apparatus. When the area specified by the specifying processing unit is the non-target area, there is another subject that has moved from the non-target area to the target area where the video of the subject was captured or to the other target area. A determination processing unit that determines the degree of similarity between the subject and the other subject based on the video acquired by the acquisition unit;
The recording processing unit, when the degree of similarity is equal to or less than a certain value, records an image of the subject with lower image quality than when the degree of similarity is larger than the certain value. Item 12. The video processing apparatus according to Item 1.   The video processing apparatus according to claim 2, wherein the recording processing unit records the video in which the subject is captured with lower image quality as the degree of similarity is lower.   When the degree of similarity is greater than the certain value, the recording processing unit increases the image quality of the image in which the subject is reflected so that the degree of similarity is less than or equal to the certain value, and the image in which the object is reflected The video processing apparatus according to claim 2, wherein:   When the area specified by the specifying processing unit is the target area, the recording processing unit is the same as the video in which the subject is reflected in the destination area and the video in which the subject is reflected in the destination area. 5. The video processing apparatus according to claim 1, wherein the video processing apparatus records the image quality. Acquiring a video imaged in an area to be imaged;
Identifying a movement source area of the subject shown in the acquired video based on map information indicating the target area and the non-target area of shooting;
When the specified area is the target area, the computer executes a step of recording a video image of the subject with a lower image quality than when the specified area is the non-target area. Video processing method. Acquire video shot in the shooting area,
Identify the area from which the subject appears in the acquired video based on the map information indicating the target area and non-target area of shooting,
When the specified area is the target area, the computer causes the computer to execute a process of recording a video image of the subject with lower image quality than when the specified area is the non-target area. Video processing program.

Images